Jump to content
Warbow

Warbow's Official ranking...

Recommended Posts

I commend the effort, if it was really you that created this model. But I don't agree with your weights. Ball St. in the top 15 is a joke. Like they say in statistics, all models are wrong, only some are less wrong than others.

Don't forget that Mid majors like ball state will get a SOS boost over SEC teams in the first half of the year because the mid major OOC is much tougher, and then over the season SEC, etc sos will improve as boise st, ball st, etc will go down.

In any computer system, the more games that are played the more accurate it gets. When SOS is such a big factor this is doubly true.

Link to post
Share on other sites

Actually, if you add up the Win Percentage and Road Win percentage you come up with 35%, the same as SOS. 35% (25% win margin & 10% road win percentage). It kinda equals things out. The 10%'s for Points Scored, Points Allowed & Win Margin separates the men from the boys.

Right, I understand, but what if the stats show that SOS explains only 5% of the likelihood of a win? The way the so-called experts do it is by gathering data on every possible variable that might predict a victory and then performing a regression analysis to assign weights. It looks something like this:

LastGameOutcome(Y) = PriorSOS(X1) + PriorWin%(X2) + PriorRoadWin%(X3) + OpponentSOS(X4) + OpponentWin%(X5) + OpponentRoadWin%(X6)

After the analysis (hypothetically):

FutureGameOutcome(Y) = 3.3(X1) + 2.4(X2) + 4.5(X3) - 1.1(X4) - 2.9(X5) - 3.9(X6)

Then you use those numbers created for each X variable as the weights in your ranking system. Otherwise, your ranking system is just based on your opinion of how much everything matters - not actual results.

Link to post
Share on other sites

Right, I understand, but what if the stats show that SOS explains only 5% of the likelihood of a win? The way the so-called experts do it is by gathering data on every possible variable that might predict a victory and then performing a regression analysis to assign weights. It looks something like this:

LastGameOutcome(Y) = PriorSOS(X1) + PriorWin%(X2) + PriorRoadWin%(X3) + OpponentSOS(X4) + OpponentWin%(X5) + OpponentRoadWin%(X6)

After the analysis (hypothetically):

FutureGameOutcome(Y) = 3.3(X1) + 2.4(X2) + 4.5(X3) - 1.1(X4) - 2.9(X5) - 3.9(X6)

Then you use those numbers created for each X variable as the weights in your ranking system. Otherwise, your ranking system is just based on your opinion of how much everything matters - not actual results.

Actually this is not a predicter of who is going to win a game. It's a ranking system based on the criteria I used that is weighed and tallied equally for every team. (I had Texas ranked over Oklahoma even before the Texas win). Because I use it to rank a teams whole season success, games played on the first week of the season is weighed actually the same as their last or privious game.

You MWC fans really crack me up. I guess when BYU pasts Boise after both teams win next weekend my ranking system will be the greatest thing in the world.

Link to post
Share on other sites
Guest ragtimeJOE

You MWC fans really crack me up. I guess when BYU pasts Boise after both teams win next weekend my ranking system will be the greatest thing in the world.

Did you have this "ranking system" last year? If so, where did Hawaii rank prior to the GA game? ---Be honest.

Link to post
Share on other sites

Yes those games count toward your win loss percentage and SOS only. Most D-1 teams have 1-AA teams on their schedule. If you play and win against a D-1AA team with a good record you get the rewards the same as if you play a D-1 team with a good record. If you play a lousy D-1AA with a losing records then your SOS suffers also.

If you took out the I-AA games, how different would the rankings look?

Link to post
Share on other sites

Actually this is not a predicter of who is going to win a game. It's a ranking system based on the criteria I used that is weighed and tallied equally for every team. (I had Texas ranked over Oklahoma even before the Texas win). Because I use it to rank a teams whole season success, games played on the first week of the season is weighed actually the same as their last or privious game.

You MWC fans really crack me up. I guess when BYU pasts Boise after both teams win next weekend my ranking system will be the greatest thing in the world.

I don't think you're quite getting it. I gave you an objective way to come up with your weights.

If you base the weights on your OPINION, then your ranking system is slightly better than worthless - regardless of where the teams show up in the rankings. If you base your weights (not rankings, the weights themselves) on the actual ability of your predictors to predict a win or loss, then your ranking system is useful to the degree that your variables can predict a win.

Also, notice that in none of my posts have I mentioned anything about where BYU, BSU, or anybody else was ranked. Just some honest feedback - not my opinion.

Link to post
Share on other sites

You MWC fans really crack me up. I guess when BYU pasts Boise after both teams win next weekend my ranking system will be the greatest thing in the world.

Sounds like a certain Hawaii fan doesn't have a football season to concentrate on?

Link to post
Share on other sites

All BCS BS aside, here's my second week of my ranking using my rankulator. Here's my criteria I use for a top 25.

35.00% SOS

25.00% Win percentage

10.00% Win Margin

10.00% Points scored

10.00% Points allowed

10.00% Away win percentage

Any ranking system that makes SOS the #1 determining factor is pure and complete garbage.

Link to post
Share on other sites

:P

Any ranking system that makes SOS the #1 determining factor is pure and complete garbage.

Maybe you should learn how to read before you throw out the garbage. I could swear that 25% (win percentage) plus 10% (away win percentage) equals to 35%. The remaining 30% has to do with how well they peformed on the field.

Sorry, didn't mean to offend any BYU fans with my high SOS percentage. Maybe I should add: Blown refs calls for 36% and a Blown refs calls against for 1%. I'm sure BYU would come out #1 after that.

Link to post
Share on other sites
On 10/13/2008 at 12:39 PM, Cougarmandias said:

Rankulator rocks!

Still rocks in 2020. Warming up the beast!

  • Like 1
  • Cheers 1
Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


  • Recently Browsing   0 members

    No registered users viewing this page.




×
×
  • Create New...